This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
Re: [patch] Repost: genattrtab speedup/memory reduction
- To: Richard Henderson <rth at redhat dot com>
- Subject: Re: [patch] Repost: genattrtab speedup/memory reduction
- From: Daniel Berlin <dan at cgsoftware dot com>
- Date: Wed, 8 Aug 2001 14:07:10 -0400 (EDT)
- cc: Jan van Male#84776 <vanmale at fenk dot wau dot nl>, <gcc-patches at gcc dot gnu dot org>, <jh at suse dot cz>
On Wed, 8 Aug 2001, Richard Henderson wrote:
> On Tue, Aug 07, 2001 at 01:54:16PM +0200, Jan van Male#84776 wrote:
> > /* Loop until nothing changes for one iteration. */
> > ! something_changed = 1;
> > ! while (something_changed)
> > {
> > ! something_changed = 0;
> > ! for (iv = insn_code_values[i]; iv; iv = iv->next)
> ----
> > /* Loop until nothing changes for one iteration. */
> > ! for (iv = insn_code_values[i]; iv; iv = iv->next)
>
> How is it that we're "looping until nothing changes",
> if the while statement has been removed?
Won't the:
while (GET_CODE (newexpr) == COND))
loop he added do this?
Before it would just skip it if it wasn't a cond, and if *anything*
changed that iteration, it wouldredo a lot. Now, it should just process it
until it's not a cond, and only need to do one iteration.
At least, I think that's the theory.
I.E. redo everything when anything changed, vs redo each thing hat
changed, until it stops changing.
>
>
> r~
>